Triton Inference Server
NVIDIA's open-source multi-framework inference engine built to maximize GPU and CPU throughput.
Explore on Agidapp Global